Upload 5 files
Browse files- config.json +67 -0
- preprocessor_config.json +12 -0
- pytorch_model.bin +3 -0
- tokenizer.json +0 -0
- vocab.txt +0 -0
config.json
ADDED
|
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_commit_hash": null,
|
| 3 |
+
"_name_or_path": "./tvp-tiny",
|
| 4 |
+
"alpha": 1.0,
|
| 5 |
+
"architectures": [
|
| 6 |
+
"TvpForVideoGrounding"
|
| 7 |
+
],
|
| 8 |
+
"attention_probs_dropout_prob": 0.1,
|
| 9 |
+
"backbone_config": {
|
| 10 |
+
"architectures": [
|
| 11 |
+
"ResNetModel"
|
| 12 |
+
],
|
| 13 |
+
"depths": [
|
| 14 |
+
2,
|
| 15 |
+
2
|
| 16 |
+
],
|
| 17 |
+
"downsample_in_bottleneck": true,
|
| 18 |
+
"hidden_sizes": [
|
| 19 |
+
64,
|
| 20 |
+
128
|
| 21 |
+
],
|
| 22 |
+
"model_type": "resnet",
|
| 23 |
+
"out_features": [
|
| 24 |
+
"stage2"
|
| 25 |
+
],
|
| 26 |
+
"out_indices": [
|
| 27 |
+
2
|
| 28 |
+
],
|
| 29 |
+
"stage_names": [
|
| 30 |
+
"stem",
|
| 31 |
+
"stage1",
|
| 32 |
+
"stage2"
|
| 33 |
+
],
|
| 34 |
+
"torch_dtype": "float32"
|
| 35 |
+
},
|
| 36 |
+
"beta": 0.1,
|
| 37 |
+
"classifier": "mlp",
|
| 38 |
+
"classifier_dropout": null,
|
| 39 |
+
"cls_hidden_scale": 2,
|
| 40 |
+
"hidden_act": "gelu",
|
| 41 |
+
"hidden_dropout_prob": 0.1,
|
| 42 |
+
"hidden_size": 128,
|
| 43 |
+
"initializer_range": 0.02,
|
| 44 |
+
"intermediate_size": 384,
|
| 45 |
+
"layer_norm_eps": 1e-12,
|
| 46 |
+
"max_grid_col_position_embeddings": 100,
|
| 47 |
+
"max_grid_row_position_embeddings": 100,
|
| 48 |
+
"max_img_size": 448,
|
| 49 |
+
"max_position_embeddings": 128,
|
| 50 |
+
"max_text_length": 20,
|
| 51 |
+
"model_type": "tvp",
|
| 52 |
+
"num_attention_heads": 4,
|
| 53 |
+
"num_frm": 8,
|
| 54 |
+
"num_hidden_layers": 4,
|
| 55 |
+
"pad_size": 96,
|
| 56 |
+
"pad_token_id": 0,
|
| 57 |
+
"position_embedding_type": "absolute",
|
| 58 |
+
"torch_dtype": "float32",
|
| 59 |
+
"transformers_version": "4.29.2",
|
| 60 |
+
"type_vocab_size": 2,
|
| 61 |
+
"use_cache": true,
|
| 62 |
+
"visual_prompter_apply": "replace",
|
| 63 |
+
"visual_prompter_type": "framepad",
|
| 64 |
+
"vocab_size": 30522,
|
| 65 |
+
"vp_apply": "replace",
|
| 66 |
+
"vp_type": "framepad"
|
| 67 |
+
}
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_center_crop": false,
|
| 3 |
+
"do_normalize": true,
|
| 4 |
+
"do_resize": true,
|
| 5 |
+
"do_rescale": false,
|
| 6 |
+
"do_padding": true,
|
| 7 |
+
"image_mean": [8.2381, 7.3115, 6.6981],
|
| 8 |
+
"image_std": [9.6335, 9.0659, 8.7213],
|
| 9 |
+
"processor_class": "TvpProcessor",
|
| 10 |
+
"padding_size": {"height": 448, "width": 448},
|
| 11 |
+
"tokenizer": "bert-base-uncased"
|
| 12 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cde4d42d12cd3f8348b237f85549eef9fcfe92421cf9364d1dab0641794077cd
|
| 3 |
+
size 32588729
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|